World Development, Vol. 19, No. 4, pp. 287-300, 1991.
Printed in Great Britain.
0305-750X/91 $3.00 + 0.00
© 1991 Pergamon Press plc
Development Narratives, Or Making the Best of
Blueprint Development
EMERY M. ROE*
University of California, Berkeley
Summary. - - The scenarios and arguments that drive and sanction much of Third World rural
development are often dismissed as myths, ideologies, conventional wis~loms or fads. Yet these
development narratives persist through time and frequently in spite of evidence learned in the
field. Instead of calling for more site-specific learning to overthrow narratives that seem to be
blueprints for development interventions across countries, the wiser course is first to examine
ways in which these narratives can be improved or superseded. Four case studies show how policy
makers and practitioners can think more enterprisingly about development narratives specifically
and blueprint development generally.
1. I N T R O D U C T I O N
No one has a good word for blueprint development. The notion that rural development can be
stenciled whole-cloth from premade plans and
blueprints has been taken to task by many wellknown critics, e.g., Hyden (1983, p. 65), Moris
(1981, pp. 19-22), Korten (1980, pp. 496-501),
and Sandford (1983, pp. 258), to name but a few.
Invariably, the remedy r e c o m m e n d e d is to abandon blueprint development in favor of a learning
process approach, one that conceives development as trial and error, where projects are
hypotheses and what is called "failure" is part of,
or should be part of, a broader learning curve
(e.g., Chambers, 1983, pp. 211-212). The comparative advantage of the learning process
approach is taken to be its flexibility and adaptability, given that the "probability of planned
actions going wrong is high in an environment
characterized by instability and uncertainty," as
is the case for many regions of the developing
world (Hyden, 1983, p. 157). Unfortunately, this
last point undermines, rather than reinforces, the
purported advantages of the learning process
approach.
The reasons we do not learn more from past
rural development efforts are precisely the same
reasons we cannot plan better for future ones. If
planning is difficult, so too must be learning, and
the performance record of rural development
points overwhelmingly in one direction: planning
has left much to be desired (Caiden and Wildavsky, 1974; Johnston and Clark, 1982, Chapter 1;
for an exception, see C o h e n and Lewis, 1987).
The preconditions for successful project and
national planning - - low environmental uncertainty, stability in goals and national objectives,
institutional memory, and redundant resources
are precisely the same preconditions for
"learning better from experience," and it is these
preconditions that are woefully lacking across
wide parts of the developing world. 1 It is the
impression of learning less and less over time
while being more and m o r e vulnerable to error
which characterizes some 30 years of subSaharan African livestock projects, for example. 2
Those caught in its clutches find little consolation
in being told that they should " e m b r a c e " error in
order to learn better (e.g., Korten, 1980, p. 498).
The learning process approach, or some more
potent remedy, could of course win the day. But
what do we do, as development practitioners, if
blueprint d e v e l o p m e n t is here to stay for the time
being? How can we make the best of a bad
situation? These are the questions addressed by
this paper. Rather than focusing on how one
might improve the learning process approach,
attention is given below to ways practitioners can
better utilize blueprint development. 3
-
-
*I am indebted to John Bruce, Bruce Johnston, David
Leonard, and Norman Uphoff for discussions on issues
raised in this article; to Esther Wangari for guiding me
to Kenya's land registration literature; and to support
from the S. V. Ciriacy-Wantrup Postdoctoral Fellowship program at the University of California, Berkeley.
The usual disclaimers apply.
287
288
WORLD DEVELOPMENT
2. F O U R D E V E L O P M E N T N A R R A T I V E S
First it must be understood that blueprint
d e v e l o p m e n t persists for precisely the same
reason said to warrant the learning process
approach. Rural d e v e l o p m e n t is a genuinely
uncertain activity, and one of the principal ways
practitioners, bureaucrats and policy makers
articulate and make sense of this uncertainty is to
tell stories or scenarios that simplify the ambiguity. Indeed, the pressure to generate narratives
about d e v e l o p m e n t is directly proportional to the
ambiguity decision makers experience over the
d e v e l o p m e n t process. The more uncertain things
seem at the microlevel, the greater the tendency
to see the scale of uncertainty at the macrolevel
to be so e n o r m o u s as to require broad explanatory narratives that can be operationalized into
standard approaches with widespread application. The unresolved failure of project blueprints
derived from d e v e l o p m e n t narratives thus often
serves only to reinforce, not lessen, the perceived
need for some sort of narrative that accounts for
the resulting increase in uncertainty.
Theso considerations raise the question of
whether rural d e v e l o p m e n t could be more productively improved by identifying ways to make
better use of those narratives that give rise to the
blueprints. To see how this might be done, four
d e v e l o p m e n t narratives are examined below.
The examples are from Africa, but could be from
anywhere in the world. Each of these narratives
has persisted in the face of strong empirical
evidence against its storyline and it is this
persistence which, the paper argues, can be more
effectively exploited. Rather than continuing to
focus on trying to undermine the narrative
evidentially, our efforts should shift to creating
and engaging counternarratives to the more
objectionable narrative or modifying that narrative to make it less objectionable. In order to
strengthen this argument, the four examples have
been chosen precisely because they appear at
first glance to demonstrate the worst in blueprint
d e v e l o p m e n t and to cry out for rectification
through a learning process approach. By arguing
that blueprint d e v e l o p m e n t can be improved
through better manipulating the narratives upon
which it is based, the examples seek to underscore that practitioners will have to deal much
more enterprisingly with a form of d e v e l o p m e n t
that persists regardless and, at times, in spite of
what is learned in the field.
A preliminary definition of " d e v e l o p m e n t narratives" stresses both their status as stories or
arguments and their differences from other
notions more familiar to the d e v e l o p m e n t practitioner, namely, ideology, myth and conventional
wisdom. The narratives discussed below follow
the c o m m o n definition of "story." Each has a
beginning, middle, and end (or premises and
conclusions, when cast in the form of an argument) and revolves around a sequence of events
or positions in which something happens or from
which something follows. 4 Typically less hortatory and normative than ideology, development
narratives tell scenarios not so much about what
should happen as about what will happen
according to their tellers - - if the events or
positions are carried out as described. Even when
their truth-value is in question, these narratives
are explicitly more programmatic than myths and
have the objective of getting their hearers to
believe or do something. In addition, the narratives, at least the four to which we now turn, are
treated by many of their tellers and hearers as
continuing to retain some general explanatory or
descriptive power even after a number of the
specific conventional wisdoms upon which they
are based are understood to be subject to serious
qualification. 5 How these and other features of
d e v e l o p m e n t narratives operate, and with what
implications, will become clearer in the
examples.
(a) The "tragedy of the commons"
The most obvious feature of the "tragedy of
the c o m m o n s " is oddly the least c o m m e n t e d on,
namely, its status as narrative. Its most famous
expositor, Garrett Hardin, goes out of his way to
tell the tragedy of the commons as a story having
all the classic properties of a beginning, middle,
and end. " T h e tragedy of the commons develops
in this way. Picture a pasture open to all . . . . "'
begins Hardin in what surely is the most quoted
passage in all of the c o m m o n property literature.
We soon are in the middle of things - - "the
rational herdsman concludes that the only sensible course for him to pursue is to add another
animal to his herd. A n d another . . . But this is
the conclusion reached by each and every
rational herdsman sharing a c o m m o n s " - - and
the end comes rapidly and palpably into sight:
"Ruin is the destination toward which all men
rush, each pursuing his own best interest in a
society that believes in the freedom of the
c o m m o n s " (Hardin, 1977, p. 20).
Reality and story, however, congrue imperfectly. When the tragedy of the commons argument is probed empirically - - for example, just
what is the evidence that desertification is caused
by overgrazing? - - the data turn out to be much
more ambiguous or outright contradictory (Sandford, 1983, Chapter 1). Even where people agree
DEVELOPMENT NARRATIVES
with Hardin that range degradation is taking
place and that many commons today are open
access free-for-alls, they often part company with
him over causes. For these critics, long-term
climatological changes along with growing and
competing land uses have led to degradation
more than the existence of the commons - - a
commons which, the critics hasten to add c o n t r a
Hardin, was often managed in a restricted access,
not open access, fashion until these exogenous
pressures undermined local management efforts
(see Panel on Common Property Resource
Management, 1986; McCay and Acheson, 1987).
Hardin merits a much closer reading than some
of his critics have given him. "It must not be
supposed that all commons are bad in all
situations," Hardin tells us, " . . .
when there
were only a few million people in the world, it
was all right to run the hunting grounds as a
commons, though even then an area was no
doubt often managed as tribal property" (Hardin, 1977, pp. 47-48). Hardin does not say that
the commons cannot be managed. The "~commons, if it is justifiable at all, is justifiable only
under conditions of low-population density"
(1977, p. 28). His pivotal point is that herders
find it to their individual advantage not to
cooperate in limiting herd numbers or ensuring
range quality even when each and every herder
recognizes that the overall stocking rate on the
commons exceeds its carrying capacity, and that
range deterioration and liveweight loss are on the
rise (Hardin, 1977, p. 72). In such a situation,
corrective measures are largely outside the initiative of the individual herder. Either the commons
has to be legislated as private property or other
coercive devices, such as taxes and user regulations, have to be instituted from the outside
(Hardin, 1977, p. 22).
In short, if we subscribe to Hardin's argument,
we should expect to find at least two states of
affairs pertaining when a rangeland tragedy of
the commons is said to exist. First, even when
herders agree that the range is in poor or already
heavily-stocked condition, they still act in a
noncooperative, competitive fashion. They
evince few if any collective practices for managing that commons, which in turn encourages its
further overutilization. Second, a tragedy of the
commons supposes that a privatized rangeland
will be better managed (e.g., have a better range
condition), than if it were a commons, other
things being equal.
The best test to date of Hardin's comprehensive version of the tragedy of the commons can
be found in a series of publications based on data
collected during the 1979-80 Botswana Water
Points Survey. 6 These data allow us to address
289
the applicability of each element of Hardin's
argument to rural eastern Botswana, an area of
the country that has repeatedly been described
by both those inside and outside the Botswana
government as in the throes of a rangeland
tragedy of the commons. First, do rural Batswana themselves perceive overgrazing to be
taking place? The evidence here is mixed. While
the range condition was found to be already at
low levels in much of the eastern communal areas
when assessed by standard range ecology
measures, a number of survey households indicated that variation in rainfall rather than livestock numbers was taken to be the major
explanation of overstocking and overgrazing
(Fortmann and Roe, '1981; Bailey, 1982). Yet
there is sufficient grounds - - in the form of low
cattle carcass weights and other interview findi n g s - to suppose that a significant proportion of
Batswana, albeit not a majority, did in fact
believe that increasing numbers of livestock were
leading to overutilization of the range in the late
1970s and early 1980s. Moreover, Botswana must
be one of the few countries in sub-Saharan Africa
that has enshrined the tragedy of the commons
argument into its national rangeland policy (see
in particular, Republic of Botswana, 1975, p. 1).
On the assumption that overgrazing is taking
place, do rural Batswana cooperate in the management of their communal resources and have
management practices to do so? An in-depth
survey by Fortmann of communal dams constructed in eastern Botswana during 1974-80
found that this was indeed the case (Fortmann
and Roe, 1981). Of the 24 dams surveyed, 21 had
some sort of collective management, be it in the
form of maintenance, regulation, or revenue
collection. In particular, all 21 dam groups had
users who jointly regulated the use of these
dams. Restrictions on numbers of users, types of
use, the manner of use, and the time of use were
found in force.
Did communal management have a positive
effect on the surrounding range condition and if
so, how did it compare to privatized resource
management? An analysis of grazing conditions
around a sample of 46 water points found that
those water points owned or managed by government or groups had better dry season range
conditions than privately-owned water points
(Roe, 1984). No one-to-one correlation was
found between private ownership, private management, and the actual restriction of livestock
watering access. Indeed, the finding that private
rights and better range condition do not go handin-hand has been confirmed on a number of
occasions in Botswana. Only three years after the
first leasehold fenced ranches were occupied
290
WORLD DEVELOPMENT
under the World Bank's First Livestock Development Project, two-thirds of them were already
overstocked (Odell and Odell, 1986, p. 7).
Conditions did not improve under the Bank's
Second Livestock Development Project. Bekure
and Dyson-Hudson (1982) found that range
management and condition on these leasehold
ranches was often no better than in the communal areas. In a related fashion, roughly 15 years
of government grazing trials undertaken periodically from the 1950s through the 1970s could
show no significant difference in range conditions
between those found under various fenced rotational systems and that observed under a
continuous "single paddock" grazing regime
approximating the communal system (see Roe
and Fortmann, 1981, p. 71; Animal Research
Production Unit, 1980, pp. 85-86). Communal
management of the dams studied in the Water
Points Survey was found not only to be ecologically efficient relative to the next best private
alternative, but cost figures indicated that this
management was economically efficient as well
(Fortmann and Roe, 1986). In short, the evidence is far from conclusive that privatization of
the Botswana commons increases the likelihood
of improving range conditions there.
These and other negative findings across the
world have been around for some time and it is
increasingly tempting to dismiss the tragedy of
the commons as some kind of old-fashioned
fable. But to do so would be misguided and
would miss the point altogether. As a development narrative, the tragedy of the commons
continues to have staying power in large part
because these negative findings and critiques in
no way dispel the chief virtue of the narrative: it
has helped to stabilize and underwrite the
assumptions needed for decision making. Policy
makers resort to the tragedy of the commons
model in order to understand what is going on
and what must be done in lieu of more elaborate
and demanding analysis, particularly when such
analysis leads only to doubts and uncertainties
about just what the story is behind rural resource
utilization. Critiques of the tragedy of the commons are doubly troublesome for the decision
maker, since they can generate rather than
reduce uncertainty. They both undermine the
assumptions of decision making and leave that
decision maker without the means to make the
transition from the discredited narrative to whatever is to replace it. In fact, the more the tragedy
of the commons is criticized and found substantively wanting, the more uncertain policy makers
can become - - why indeed did Batswana manage
their water points collectively'? - - and the more
pressure they feel to hold onto the simple
heuristics they have, no matter how worn around
the edges they now appear. In short, a critique,
like that based on the Botswana data, never tells
its own story - - its point-by-point rebuttal does
not have its own beginning, middle, and end - and often serves only to raise doubts that the
critique itself cannot answer. 7
What displaces a development narrative are
not just the negative findings that seem to refute
it. Displacement also requires an equally
straightforward narrative that tells a better story.
The appeal of the tragedy of the commons to
livestock rangeland project designers has been its
blueprinted design implications for privatizing
the commons and legislating stock controls. If
project designers are to reject the blueprint, they
must have another story whose design implications are equally as obvious to them. The
operating assumption here is that if decision
makers are to move beyond the prevailing model
of an entirely unmanaged and open-access commons, they will do so not merely by being told
that reality is more complex than has been
thought, but also by having a counternarrative
which can predict when common property management will take place or not and what are the
implications of either event.
Moreover, the counternarrative will have to be
as parsimonious as the tragedy of the commons
argument, but comprehensive enough to explain
not only when management of the commons
occurs, but when the tragedy takes places instead. Indeed, the ideal counternarrative for the
project designer and policy maker would be like
the tragedy of the commons in having to rely on
nothing more sophisticated than introductory
microeconomics (for an example of one such
model, see Roe, 1987a). While such a conclusion
will offend many social scientists, its dismissal is
surely premature, as we all wait for a counternarrative - - any counternarrative
to the
tragedy of the commons that is more substantial
than the critiques offered up so far.
(b) Land registration and increased agricultural
productivity
For at least the past 35 years, one of the most
potent development narratives in Kenya has
been that land registration leads to increased
agricultural productivity. Once land is adjudicated, consolidated and registered, so this argument goes, the landowner will be in a position to
use the title deed as collateral for securing credit
with which to invest in improving and intensifying agricultural production on the land concerned. Dating from (if not before) the blueprint
laid out in the government's 1954 Swynnerton
DEVELOPMENT NARRATIVES
Plan (Colony and Protectorate of Kenya, 1954,
pp. 8-9), the argument remains extremely popular among Kenyan politicians, senior civil servants and social scientists (e.g. Haji, 1989, p. 20
and Mutiso, 1988, p. 48).
Empirical studies have repeatedly failed to find
a positive causal link connecting the government's land registration program to expanded
credit opportunities and thereby to increased
agricultural productivity. Over the years, the
effects of land registration in one district (Embu)
have been studied in detail by different researchers, while others have undertaken pointin-time research on the same topic - - much of it
in the form of household surveys at the farm level
for localities in at least 13 other districts
(Meru, Nyeri, Kiambu, Kwale, Kisii, Murang'a,
Taita-Taveta, Kisumu, South Nyanza, Nakuru,
Kericho, Machakos, and Kakamega) covering
much of the country's most agriculturally productive cropland. All the studies have failed to
confirm or have raised serious doubts about the
scenario linking land registration to agricultural
production (Sorrenson, 1967; Barber, 1970; Bernard, 1972; Gray, 1972; Wilson, 1971; Coldham,
1979; Hunt, 1984; Njeru, 1978; Brokensha and
Njeru, 1977; Haugerud, 1981, 1983, and 1989;
Shipton, 1985; Odingo, 1985; Green, 1985;
Fleuret, 1988; see also Wangari, forthcoming;
and Olouch-Kosura, forthcoming), s This general
finding is made all the more significant by the
equally demonstrated interregional ethnic and
socioeconomic diversity of the country's rural
households.
In contrast to the narrative, the cumulative
picture left by the research suggests that once
landowners are registered, many do not bother to
obtain their title deeds (they would never risk
losing them on anything as uncertain as loan
defaults); of those landowners who do obtain
titles, not all of them do so to obtain credit (they
may have to sell their land or parts of it to meet
school fees and other household expenses); of
those who want to use their titles to obtain loans,
not all actually receive credit (farmers may not
know where to go for credit or meet other
requirements of the lending institution, which in
turn might not have the funds to lend); of those
landowners who actually succeed in using their
titles for securing credit, a number of them use
the loans for nonagricultural investments (e.g.,
their off-farm businesses); once the registered
landowner dies or sells off parcels, the new
landowner frequently does not reregister; and
those who do not reregister or who could not
legally register in the first place - - mostly women
- - are ineligible for title-secured loans.
Nor has the problem only been one of a low
-
-
291
conditional probability that, once registered,
credit will be obtained and agriculture intensified. In some cases, land registration and increasing agricultural production may actually be
negatively related. When registered, some landowners feel (i) they can leave the land idle
without fear of someone else claiming it because
no one was planting there; (ii) they have the
"freedom" to sell land without real consent from
those dependent on it and whose labor makes it
as productive as it is; and (iii) they can enter land
transactions for speculation purposes only. As a
result of these and other factors, several authorities have concluded that Kenya's land registration program has ihcreased insecurity, rather
than the security, of tenure in many parts of the
country. 9
Yet, the recurrent finding that registration
does not increase production via the credit
mechanism has not changed one iota the belief of
many respected Kenyans that registration has a
positive and widespread effect on agriculture.
What then is the policy maker to do if he or she
feels compelled to pursue the topic of land
registration's supposedly positive effect on agricultural productivity? One could, of course,
continue to analyze the subject of agricultural
credit on the assumption that land registration
would intensify agriculture if only credit is made
more timely, convenient, and adequate to more
smallholders. Pushing credit, however, has continually proven to afford about as much leverage
in Kenya's agricultural sector as pushing string
does elsewhere (e.g., World Bank, 1985).
Another option would be to explore the other
links between registration and agriculture that
appear to have potential in offering up a counternarrative relevant for policy and program
development. Land registration, for example,
seems to have affected shifts in crop mixes, land
concentration and fragmentation, and the utilization of both nonfarm income and credit for
nonagricultural purposes. Nevertheless, if past
research is any indication, this approach will
yield mixed signals for the policy maker. Land
concentration has boosted agricultural productivity in the view of one study, while another finds
otherwise; fragmentation is an ecologically valid,
risk-averse response of farmers to some researchers, while others focus on what they see as
increasingly subeconomic holdings (e.g., see
Shipton, 1985; Haugerud, 1983). The policy
maker who chooses this option, as noted earlier,
will have to balance the findings of increasing
complexity at the microlevel with the widening
scale of Kenya's land problems that seem to
demand standard approaches to their management and amelioration.
292
WORLD DEVELOPMENT
A very different option suggests itself, however, if the operating assumption is that both the
popular narrative linking registration to production and its blueprint, the government's land
registration program, will persist for the foreseeable future in the absence of any viable alternative and regardless of empirical findings that
erode their credibility. Assuming this is so, the
question then becomes one of focusing the policy
maker's attention on those few topics where land
registration offers some promise of actually
expanding agricultural production.
One prospect stands out in its priority and
scale - - estimating the extent to which the
implementation of a progressive land tax in
Kenya could intensify agricultural production by
discouraging land speculation, absentee management, uneconomic fragmentation, and nonproductively large holdings. The records of the land
registration program as to who is registered or
holds title deed will greatly facilitate the operation of such a tax, if and when it is introduced.
Moreover, it is difficult to conceive of a more
efficient way to up-date these records for the
unregistered subdivisions mentioned earlier, i.e.,
the levy could serve as the necessary incentive for
the reregistration of land currently registered in
the names of other owners, since the tax would
presumably be assessed on those whose names
show up in the records. Even though land
taxation is controversial in Kenya (albeit the
government is proposing to introduce a tax on
agricultural production), a focus on the promising positive links between registration and agriculture has the considerable merit of being
consistent with the development narrative that
has hitherto resisted all manner of assaults on it.
Indeed, the narrative's blueprint becomes one
way of altering rather than displacing the narrative itself. The existence of a huge and growing
ministry, staff and budget - - all committed to
ensuring that land registration is the blueprint
across the country - - will inevitably increase
pressure to justify their purported programmatic
effect on intensifying agriculture, even if the
specific mechanism for linking registration and
production is no longer principally that of titlesecured credit, m
(c) Systems analysis and sectoral integration
Systems thinking suffuses the practitioner's
approach to rural development. The field of
livestock rangeland development provides an
especially good illustration of the development
narrative in question. For example, few specialists would quarrel with the following:
Pastoral production is normally and correctly part
of wider production systems; changes in any element have ramifications in all others (Galaty and
Aronson, 1981, p. 21).
Livestock production is a very complex system
which has many interrelated components such as
climate, soil, plants and obviously animals operating with a high degree of interaction within a certain
economic and social environment (Sidahmed and
Koong, 1984, p. 61).
The fact that pastoralism needs to be seen in a
regional perspective rather than an isolated production system for an understanding of the changes in
living conditions of pastoralists has been pointed
out (Hjort, 1981, p. 135).
But why stop at "region"? The "system" entails,
after a point, supraregional levels in a long chain
of behavioral links which
starts with land managers and their direct relations
with the land ( . . . stocking densities.., and so on).
Then on the next link concerns their relations with
each other, other land users, and groups in the
wider society who affect them in any way, which in
turn determines land management. The state and
the world economy constitute the last links in the
chain (Blaikie and Brookfield, 1987, p. 27).
Once something has evolved into such a longlinked "system," description frequently becomes
prescription. "A local systems approach holds
the best hope of success in development for
pastoralists,'" recommend two social scientists
(Galaty and Aronson, 1981, p. 20). "The Panel
recommends a systems approach to all phases of
project planning and execution," advises a US
Agency for International Development taskforce
on rangeland/livestock projects (USAID, 1985,
p. 7). For Sidahmed and Koong the moral of
their systems approach is clear, i.e., "any
attempts towards [livestock] development should
be preceded by the construction of a mathematical model which should contain all the essential
elements of the current production system"
(1984, p. 74).
As this last statement should indicate, making
a systems critique can be easier than living with
its implications. The development syllogism in
the critique seems inescapable: integrated system, integrated intervention. As many readers
already know, a number of integrated rural
development programs (particularly in the agricultural sector) have been "disintegrating agricultural development" (Leonard, 1984). Ministerial
portfolios, based often on apparent sectoral
divisions in the political economy, seem as a rule
not to accommodate the need for intersectoral
coordination and integration. Moreover, once
one accepts the validity of a narrative that posits
DEVELOPMENT NARRATIVES
a long chain of putative causality between the
international economy and the local herder, the
probability of finding something wrong along the
way increases exponentially. Critics continually
find projects that are local successes, but system
failures, in a world where localized interventions,
such as projects, provide little leverage in
correcting what are perceived to be systemic
dysfunctions. Nothing seems to work right because the conditional probability of doing so
approaches the vanishing point. Development
syllogism has thus become development tautology.
Yet the practitioner who objects to the systems
approach to rural development runs the risk of
always having to recant the objection. Those who
do not believe that cause gives way to effect
which gives way to cause, etc., threaten to end up
like the recanting Bishop Cranmer at the stake,
saying, as he stretched his writing hand into the
rising flames, "This hand hath offended" and it
shall be the first to burn. After all, what rural
development practitioner is foolish enough to
operate as if cause and effect had been suspended?
One of the more recent and salutary developments in the systems approach has been to
introduce the distinction between tightly and
loosely coupled systems. Many early integrated
rural development projects were clearly designed
as tightly coupled systems, i.e., everything had to
occur in a sequence of steps, there was only one
way to achieve to desired objective, little or no
slack existed in the project, and any delays or
missteps could paralyze activities throughout the
rest of the project cycle. What has not been
sufficiently observed, however, is the loosely
coupled nature of many rural development systems that set the context in which project cycles
have to operate, i.e., environments in which
delays have to be accommodated, the order of
sequences often chan]ge, and where slack seems
particularly evident. 1" For example, Moris and
Thom argue that irrigation scheme managers in
many parts of Africa must contend with what is
best characterized as a loosely coupled system
where a myriad of agencies, levels of authority,
and formal as well as informal networks of
communication are relevant to the manager's
activities, but not coordinated for the purposes of
facilitating his or her management (1987, pp.
430-431).
In fact, many rural development processes are
best understood as a m i x of tightly coupled and
loosely coupled systems. A livestock rangeland
example will again help illustrate how this is so.12
Research in Botswana found three tightly
coupled levels of water use and management
293
governing any given rural water point: the site
immediately surrounding the water point, the
locality in which the water point is found, and the
compound locality in which the water point is
located (that is, the set of different localities over
which the users of that water point typically
reside and work during the year).
When rural Batswana are physically at the
water point site, they often are keenly aware of
the physical condition of adjacent land as well as
the type of water source in question (e.g.,
boreholes, because they are mechanized, are
operated differently than open wells). At the
locality level, how a water point is managed and
used is in turn affected by, among other things,
the availability of labor for fetching water, which
varies by locality (some household members
move to the cropping fields for planting, while
others, such as children who would otherwise
fetch water, remain in 'village schools); by the
locality's topography and hydrogeology; and by
the locality's prevailing land uses (e.g., in Botswana, villages and grazing areas are typically
dominated by borehole development, while
mixed cropping and grazing areas frequently
have a greater variety of sources). Equally
important, the operation of a specific water point
has to be seen within the context of the availability and accessibility of alternative water points in
the locality.
Because members of rural households in Botswana have frequently shifted their household
compounds over the course of the year - - in the
cropping season, they have gone to the "lands"
where their field homes are and, after harvest,
they have returned to their village homes for the
rest of the year - - one finds that the demand and
supply of water has shifted as well over the
course of the year across the localities concerned.
It is at the compound locality level that the
specific water point in question frequently has to
be seen as one of a sequence of water points over
time and space. Households fall back from many,
often surface, water sources in the wet season on
their lands to fewer groundwater sources in the
home village during the dry season.
These three levels are tightly coupled when
compared to those larger regional, national, and
international domains pertinent to rural water
use and management. The specific water point is
at a site, in a locality, and within a compound
locality at the same moment. Indeed, the three
levels are able to be identified precisely because
they share the same unit of analysis at the same
time. This is not true for the other levels. No one
can doubt national considerations affect rural
water use, e.g., Botswana relies on boreholes
much more heavily than, say, does Kenya in
294
WORLD DEVELOPMENT
many of its arid and semi-arid lands ( A S A L s ) .
Similarly, international considerations clearly
affect water use conditions in a given county.
E u r o p e a n E c o n o m i c C o m m u n i t y ( E E C ) price
subsidies on beef exports - - which Botswana has
taken advantage of, but which Kenya has not
because of its relatively weaker veterinary control efforts - - helped make livestock investment
in Botswana a much more profitable enterprise
than it has been in the comparable A S A L s of
Kenya. Yet, when moving to these higher levels,
it becomes increasingly difficult to keep the
specific water point as the unit of analysis fixed in
view by the three more local levels. Just how
Botswana's being a borehole culture and the
beneficiary of international price subsidies affects
one of the thousands of open wells in eastern
Botswana is doubtless not clear to the average
user, let alone to the outside observer. These
other levels are, in short, connected to rural
water use and m a n a g e m e n t , but in a loosely
coupled fashion when compared to the tightly
coupled processes governing the operation of any
specific water point.
What such discriminations allow one to do is to
adhere to the narrative of rural d e v e l o p m e n t as a
long-linked system stretching from the local to
the international, but to recognize that some
systems are so loosely coupled that practitioners
can treat them as if they were not systems at all.
From the standpoint of any single herder (like
that of any specific water point), the health sector
probably is not "'integrated" with the agricultural
sector which in turn is not integrated with the
wage s e c t o r . . , no matter what the government
planners say or believe or feel should be the case.
Indeed, much of the failure of tightly coupled
integrated rural development projects becomes
perfectly intelligible the m o m e n t we assume that
for the typical project beneficiary "sectors" arc
not integrated at all, or at least not at the local
level(s) relevant to the person concerned. Ironically, this is precisely the reason "'multisectoral'"
programs are still needed in parts of the developing world, not because project activities are
tightly coupled and functionally integrated at the
sectoral level, but because multisectoral projects
are often in practice loosely coupled and are
better able to adapt to the "unintegrated" nature
of development. They have had more flexibility
at the local level in undertaking different projects
over different geographical ureas and better
chances at exploiting new project opportunities
as they arise - - both major advantages in an
environment typically characterized by the uncertainty of multiple actors, objectives, and
criteria for evaluating program performance
(Roe, 1985; see also Cohen, 1987, Chapter 7).
(d) Repetitive budgeting by national
governments 13
Disarray in budgeting by governments at the
national level is patently evident. Neither the
developing world nor the developed countries
can cope budgetarily. "'How we Americans used
to deride the 'banana republics" of the world for
their 'repetitive budgeting' under which the
budget was reallocated many times during the
year, until it became hardly recognizable, truly a
thing of shreds and patches . . . . Yet resolutions
that continue last year's funding for agencies, for
want of ability to agree on this year's, are
becoming a way of life in the United States," as
one long-time c o m m e n t a t o r on budgeting put it
(Wildavsky, 1984, p. 252: see Schick, 1988 for the
O E C D countries).
It is in the developing world, however, where
repetitive budgeting has received the greatest
attention. U n d e r this kind of budgeting, according to Caiden and Wildavsky's Planning and
Budgeting in Poor Countries, the government
budget is not made once and for all whcn estimates
are submitted and approved; rather, as the process
of budgeting is repeated, it is made and remade
over the course of the y e a r . . . The entire budget is
treatcd as if each item werc supplemental, subject
to renegotiation at the last minute . . . Repetitive
budgeting . . . is found in most poor countries. Its
most extreme manifestation is as cash flow budgeting where changes may be made from day to day or
cven from one hour to the next (1974, pp. 71 72).
In practice, the presence of repetitive budgeting
is best measured by the gaps that can perforate
the national budgetary process. For example,
there have in the past been substantial differences between what is printed in the government
of Kenya's five-year national development plan
and what the government eventually budgets m
its three-year forward budget, between that
printed forward budget and what has been
budgeted annually in the published estimates,
between what the ministries formally requested
to have budgeted in the annual estimates and
what ended up being allocated to them by way of
official Treasury warrants, and between that
allocation and what audit reports subsequently
show was actually spent (Roe, 1986: for similar
gaps in the US federal budgetary process, see
Caidcn, 1984). Accordingly, a very large gap can
exist between what was originally planned and
what is eventually implemented, and it is this gap
which motivates many of more critical accounts
on the failure of large-scale government programs in both the developing and developed
worlds.
The facts, however, do not always accord with
DEVELOPMENT NARRATIVES
this universalized narrative linking repetitive
budgeting at the national level to poor project
implementation at the local level. It simply is not
the case, for example, that sub-Saharan governments all budget or implement alike, notwithstanding blanket terms like "the crisis in Africa."
From what accounts we have, it appears the
government of Botswana has budgeted differently than, say, Kenya, whose repetitive budgeting
in turn has not been as severe or disabling as it
has been in Nigeria (Roe, 1988d; Omolehinwa
and Roe, 1989). Similarly, as difficult as budgeting has been in Kenya or Nigeria, they have at
least been able to produce yearly budgets in
contrast to some more politically troubled governments (e.g., Angola; see Lamb, 1984, p. 175).
Such discriminations should be terribly important
for those who believe that rural development
depends in some significant sense on government
budgets.
More empirical work on these intercountry
differences is needed and would certainly go
some way in undermining the grim narrative's
depiction of African national budgeting. In the
absence of these case studies, are there other
ways to weaken the narrative? Several options
have been suggested in the previous sections. In
theory, one could try to create a counternarrative; use the blueprints generated by the narrative to alter it; or fill in the "details" of the
narrative in order to make it less misleading.
There are probably many cases, however, where
the practitioner has little leeway to be creative.
The best he or she may be able to do is to engage
another already existing narrative, which, once
engaged, conflicts with the more objectionable
one.
A US example illustrates how this has been
done. 14 The California Department of Motor
Vehicles (DMV) has had a reputation for very
long lines and very slow service. This narrative of
a government department paralyzed by red tape
and inaction was and still is to some extent very
popular in the press and the public's mind. Senior
DMV managers have responded by engaging two
other pre-existing narratives that are also
accepted in the United States and elsewhere,
namely, the widespread scenarios about how
"computerization improves bureaucratic efficiency" and how "personalized service is what really
makes bureaucracies effective." In this case, the
managers had DMV field operations computerized and initiated a program whereby service
by appointment could be provided to those who
formerly had to wait in line. The issue here is not
whether these other narratives proved accurate
- - in this example computerization and service
by appointment did lead to some improvement
-
295
but that their storylines ran against the
dominant narrative of a government bureaucracy
mired in its own indifference to the needs of the
people it is meant to serve. One may be tempted
to think of the actions of these senior managers
as "public relations," but to do so would deny the
highly circumscribed nature of the public arena in
which they had to make decisions. That the
managers maneuvered at all is a credit to them.15
One can see parallel developments in government budgeting overseas. Computerization of
the government of Kenya's budget was introduced when budgetary and financial management there were widely considered to be at a low
point. Pinckney, Cohen and Leonard recount
how the introduction of microcomputers into the
Ministries of Agriculture and Livestock Development improved the efficiency and effectiveness of
scarce, skilled managers in their budgetary process (1983, pp. 151 and 167). They go on to add
that "the presence of the microcomputer visibly
changed the technology of financial decisionmaking . . . [It] legitimized a r61e in financial
decision-making for those who could use it,
therefore making it easier for new, reformminded individuals to participate actively" (Pinckney, Cohen and Leonard, 1983, p. 166). Further
legitimation came from the World Bank in the
form of one of its elect "box" case studies for the
1988 World Development Report: "Overall the
Ministry of Agriculture developed better management tools and information systems, aided by
the introduction of microcomputers. The overall
quality of agricultural programs improved
markedly during this period" (World Bank, 1988,
p. 129). This technology subsequently became an
important part of the Treasury's budget preparation exercise and the government estimates for
1985-86 were the first to be produced by microcomputer (Peterson, 1988). Once again, the
impression, certainly among members of the
donor community, has been that this extension in
the use of microcomputers has also increased the
budgetary and financial management capacity of
Treasury officials and advisors (see Wescott,
1986). While the government of Kenya's budgeting is still acknowledged to have problems,
reforms such as the introduction of microcomputers have certainly helped alter its image of a
repetitive budgetary process that cannot be
improved.
-
3. CONCLUSION
At this point it should be clear that blueprint
planning and a learning process need not be
mutually exclusive, at least to the extent the
296
WORLD DEVELOPMENT
latter o p e r a t e s within the context of the former.
C o u n t e r n a r r a t i v e s to the tragedy of the comm o n s , for e x a m p l e , can a n d have b e e n b a s e d on
what has b e e n l e a r n e d in the field a b o u t c o m m o n
p r o p e r t y resource m a n a g e m e n t ( R o e , 1987a).
Similarly, no o n e is arguing t h a t since b u d g e t
c o m p u t e r i z a t i o n seems to have w o r k e d in Kenya,
it should be copied in Bolivia in the absence of
learning to what extent, if at all, " c o m p u t e r i z a tion i m p r o v e s b u d g e t i n g " is even a d e v e l o p m e n t
n a r r a t i v e there. T h e a r g u m e n t here has not b e e n
that b l u e p r i n t d e v e l o p m e n t that suits o n e c o u n t r y
should suit t h e m all, b u t r a t h e r a m o r e m o d e s t
claim. B l u e p r i n t d e v e l o p m e n t can be m a d e more
suitable to the n e e d s of a given c o u n t r y by first
learning how to b e t t e r m a n i p u l a t e the developm e n t n a r r a t i v e s that s u p p o r t these blueprints.
Such a conclusion, h o w e v e r , should not be t a k e n
to m e a n t h a t d e v e l o p m e n t n a r r a t i v e s are to be
e v a l u a t e d solely on the g r o u n d s of how well
learning i n t r o d u c e s a m e a s u r e of realism into
t h e m . In the four cases discussed a b o v e , it is not
a q u e s t i o n of which n a r r a t i v e s are m o r e accurate
or have g r e a t e r verisimilitude, lo
It m a y be best to conceive of m a n y developm e n t n a r r a t i v e s as functionally e q u i v a l e n t to the
early a n i m a t e d c a r t o o n c o n v e n t i o n that animals
have four instead of five fingers. T h e c o n v e n t i o n
tells us at once that these c a r t o o n figures are
s u p p o s e d to be m o r e "like" us t h a n not - - after
all they have fingers - - without, h o w e v e r , b e i n g
us just b e c a u s e they o t h e r w i s e b e h a v e as we do.
So too for d e v e l o p m e n t narratives: they tell us at
once t h a t things h a p p e n "like" the way they are
described I
after all n a r r a t i v e s relate things
causally I without, h o w e v e r , reflecting the fact
that things h a p p e n in rural d e v e l o p m e n t so
uncertainly. N a r r a t i v e s m a y be caricatures of
reality, but t h e r e is no p r e t e n s e otherwise for
m a n y of t h e m . Believing a n a r r a t i v e a n d acting
upon it is just like w a t c h i n g a story unfold on the
screen a n d b e i n g m o v e d by it in the same m a n n e r
as o n e is m o v e d by real events. I n d e e d , w h e n o n e
story m o r e t h a n any o t h e r b e c o m e s the way we
best articulate o u r " r e a l " feelings or m a k e sense
of the u n c e r t a i n t i e s and ambiguities a r o u n d a n d
in us, t h e n the force of the n a r r a t i v e in q u e s t i o n
b e c o m e s compelling. Obviously, the day rural
d e v e l o p m e n t b e c o m e s less u n c e r t a i n and riddled
with d i s a g r e e m e n t is the day we can begin to
dispense with some of the d e v e l o p m e n t narratives t h a t p o p u l a t e o u r profession. Until t h e n ,
rural d e v e l o p m e n t practitioners must t h i n k m o r e
positively t h a n has s o m e t i m e s b e e n the case
a b o u t h o w to m a k e the best of what we already
have, m e a g e r a n d t r o u b l i n g as it often is.
T h e a b o v e r e m a r k s have b e e n d e v o t e d to a
class of d e v e l o p m e n t n a r r a t i v e s that have, by and
large, r e m a i n e d n a r r a t i v e s only. W h a t deserves
m u c h more a t t e n t i o n is why a n d how a few of
these n a r r a t i v e s cease to be just that a n d b e c o m e
plausible assertions. Plausible assertions are
those d e v e l o p m e n t n a r r a t i v e s that can be justified as applicable to a site on the basis of long
e x p e r i e n c e and o b s e r v a t i o n . T h e y are m o r e than
the latest i m p o r t e d d e v e l o p m e n t fad, but m u c h
less t h a n an empirical generalization, let alone a
testable hypothesis. P e r h a p s the best k n o w n
plausible assertion in the rural d e v e l o p m e n t
literature, and o n e that started out as a developm e n t narrative, is the scenario revolving a r o u n d
how road c o n s t r u c t i o n in an area can greatly
facilitate the s u r r o u n d i n g e c o n o m i c growth. T h e
assertion clearly c a n n o t be generalized to all
d e v e l o p i n g c o u n t r i e s or to all regions in some
countries. Nor is it a hypothesis that can always
be tested, since the scenario is very f r e q u e n t l y
cast in the past tense, i.e., had t r a n s p o r t a t i o n
infrastructure b e e n c o n s t r u c t e d t h e n , it wonld
have led to m o r e d e v e l o p m e n t later. Nor is the
assertion really based on learning in the sense of
trial and error. It is plausible solely because very
different p e o p l e would n o n e t h e l e s s be able to
i n t e r r o g a t e and d e f e n d the scenario on the basis
of t h e i r detailed i n v o l v e m e n t , knowledge and
familiarity with the area in question, notwiths t a n d i n g the a b s e n c e t h e r e of any m a j o r road
c o n s t r u c t i o n p r o g r a m in the past. Case studies on
the tangled a n d highly selective process w h e r e b y
o n e of m a n y d e v e l o p m e n t narratives is transf o r m e d into a plausible assertion are a m a t t e r of
priority, if not urgency, r
NOTES
1. Yct the pressures for long-term planning remain
strong, at least in Africa, though for a different mix of
reasons that pertained when Caiden and Wildavsky
wrote their classic dissection of the planning mystique
in Planning and Budgeting in Poor Countries. Today,
the persistence of structural adjustment and budget
ceilings are reinforcing factors. The response of African
governments to the series of IMF/World Bank structural adjustment programs has been to call for longerterm planning horizons to counter what they see as the
invariably short-term and short-sighted corrections
suffered under SAPs (see, for example, the recent
United Nations Economic Commission on Africa,
1989, p. 33). Moreover, the increased use of ceilings on
ministry budgets as a form of year-to-year expenditure
control (for an example, see Roe, 1986) has fitted in
quite neatly with those who argue that a intermediateto long-term macroeconomic framework is needed in
part to establish just what these yearly ceilings should
be (e.g., see Ramakrishnan, 1989, pp. 2, 26-27).
2. For a discussion of this phenomenon as it is found
in irrigation projects, see Roe (1988a).
DEVELOPMENT NARRATIVES
3. Those who may doubt the vitality of blueprint
development need only consider one of many instances
of its continuing appeal. A recent United Nations'
book, for example, regrets that the Lagos Plan of
Action "does not provide a blueprint of the actual
policies that are required for its implementation at the
country level" (Kawamura, 1989, p. 270). Another UN
publication of the same year however refers to the Plan
of Action as "Africa's premier blueprint for development" (United Nations, 1989).
4. This definition of narrative and story is a standard
one (e.g., see the respective entries in Prince, 1987).
5. How the truth of a narrative does not necessarily
derive from or depend upon the truth of its constituent
parts is discussed in Roth (1989), pp. 456ff.
297
Government does not have a limitless source of funds
and, therefore, it is incumbent upon us all to optimise
the utilization of the scarce resources at our disposal"
(Ministry of Lands and Settlement, 1988).
11. The sociologist, Charles Perrow, provides an
excellent description of these general properties of
tightly and loosely coupled systems (Perrow, 1984, pp.
93-96).
12. More details on this illustration can be found in
Roe and Fortmann (1982, Chapter VI).
13. The following two paragraphs rely heavily on Roe
(1988c).
14.
The example comes Sullivan (1987).
6. The Survey was funded by the Government of
Botswana's Ministry of Agriculture and USAID
through the Center for International Studies, Cornell
University.
15. For a discussior~ of the difficult and different
contexts in which government officials administer and
manage in the developed and developing worlds, see
Roe (1988b).
7. The asymmetrical relationship between public
narratives and their critiques is explored in Roe (1989).
The analysis of public policy narratives in the United
States is further explored in case studies on the
controversies over irrigation-related toxicity and salinity in the San Joaquin Valley, animal rights, nuclear
reactors not being built according to design, and the
Cuban missile crisis in Hukkinen, Roe and Rochlin
(forthcoming) and Roe (forthcoming. a, forthcoming.
b).
16. This is not to argue against criticizing development narratives for their lack of realism. For examples
of conventional wisdoms, puzzles, myths and folktales
in rural development that deserve much greater scrutiny, see Roe (1987b, 1988e, 1989b, and 1989c).
8. See Green (1985) for a more detailed review and
bibliography of the literature on Kenya's land registration program.
9. For related comments on contemporary insecurity
of land tenure in Kenya, see Collier and Lal (1986,
p. 131).
17. An important element of this process would seem
to be what makes for "good advisors" to a government.
For a typology and discussion of such advisors, see Roe
(1988f). A specific discussion on how some "strategic
notions" in agricultural development (a term having
affinities to that of development narratives) have
become more compelling over time can be found in the
draft manuscript of Bruce F. Johnston, Peter Kilby and
Thomas P. Tomich, tentatively titled "The Political
Economy of Agricultural Development and Structural
Transformation."
REFERENCES
10. That the ministry responsible for Kenya's land
registration program is under increased pressure to
better justify its expenditures is evidenced by a recent
memo from the ministry's permanent secretary to his
staff: "During the past six months or so, His Excellency
the President has on a number of occasions impressed
upon this Ministry the need to expand and intensify our
work in land surveying, land adjudication and land
administration so that the desired goals in land tenure,
and the attendant social and economic benefits, can be
realized throughout the Republic in the shortest time
possible . . . .
A review of the performance of this
Ministry over the last few years indicates clearly that
our performance and output have been below expectation and that drastic changes are needed in our
operational strategies, priorities and personal attitudes
if we are to fulfil the mammoth responsibilities entrusted to us by the Government . . . .
While it is
acknowledged that external factors such as scarcity of
funds, personnel and equipment may retard the implementation of the [ministry's work] programme in
certain cases, it must also be recognised that the
Animal Production Research Unit, Beef Production
and Range Management in Botswana (Gaborone,
Botswana: Ministry of Agriculture, 1980).
Bailey, Charles, Cattle Husbandry in the Communal
Areas of Eastern Botswana, PhD dissertation
(Ithaca, New York: Cornell University, 1982).
Barber, William, "'Land reform and economic change
among African farmers in Kenya," Economic Development and Cultural Change, Vol. 19, No. 1
(October 1970), pp. 6-24.
Bekure, Solomon, and Neville Dyson-Hudson, The
Operation and Viability of the Second Livestock
Development Project (1497-BT): Selected Issues,
Report (Gaborone, Botswana: Ministry of Agriculture, 1982).
Bernard, Frank, East of Mount Kenya: Meru Agriculture in Transition (Munich: Weltforum Verlag,
1972).
Blaikie, Piers, and Harold Brookfield, Land Degradation and Society (London: Methuen, 1987).
Brokensha, David, and E. H. N. Njeru, "Some
298
WORLD DEVELOPMENT
consequences of land adjudication in Mbere Division, Embu," Working Paper No. 320 (Nairobi:
Institute of Development Studies, University of
Nairobi, 1977).
Caiden, Naomi, "The new rules of the federal budget
game," Public Administration Review, Vol. 44, No. 2
(March/April, 1984), pp. 109-118.
Caiden, Naomi, and Aaron Wildavsky, Planning and
Budgeting in Poor Countries (New York: John Wiley
& Sons, 1974).
Chambers, Robert, Rural Development: Putting the
Last First (London: Longman, 1983).
Cohen, John, Integrated Rural Development: The
Ethiopian Experience and the Debate (Uppsala,
Sweden: The Scandinavian Institute of African
Studies, 1987).
Cohen, John, and David Lewis, "'Role of government
in combating food shortages: Lessons from Kenya
1984-85,'" in Michael Glantz (Ed.), Drought and
Hunger in Africa (Cambridge: Cambridge University
Press, 1987).
Coldham, Simon, "'Land-tenure reform in Kenya: The
limits of the law," The Journal of Modern African
Studies, Vol. 17, No. 4 (December 1979), pp. 615627.
Collier, Paul, and Deepak Lal, Labour and Poverty in
Kenya: 1900-1980 (Oxford: Clarendon Press, 1986),
Colony and Protectorate of Kenya, A Plan to Intensify
the Development of African Agriculture in Kenya
(compiled by R. J. M. Swynnerton, Nairobi:
Government Printer, 1954).
Fleuret, Anne, "Some consequences of tenure and
agrarian reform in Taita, Kenya," in R. E. Downs
and S. P. Renya (Eds.), Land and Society in
Contemporary Africa (Hanover, New Hampshire:
University Press of New England, 1988).
Fortmann, Louise, and Emery Roe, "'Common property management of water in Botswana," in Proceedings of the Conference on Common Property
Resource Management (Washington, DC: National
Academy Press, 1986).
Fortmann, Louise, and Emery Roe, The Water Points
Survey, Report (Gaborone, Botswana: Ministry of
Agriculture and Center for International Studies,
Cornell University, 1981).
Galaty, John, and Dan Aronson, "'Research priorities
and pastoralist development: What is to be done?" in
John Galaty, Dan Aronson, Philip Salzman, and
Amy Chouinard (Eds.), The Future of" Pastoral
Peoples: Proceedings of a Conference Held in Nairobi, Kenya, ~ August, 1980 (Ottawa: International
Development Research Center, 1981).
Gray, Nancy, "Acceptance of land adjudication among
the Digo,'" Discussion Paper No. 37 (Nairobi:
Institute of African Studies, 1972).
Green, Joy, "Evaluating the impact of consolidation of
holdings, individualization of tenure, and registration of title: Lessons from Kenya," LTC Paper No.
129 (Madison: Land Tenure Center, University of
Wisconsin, 1985).
Haft, Yusuf, "'Focus on 1989 Nakuru show - - Rift
Valley: Land of plenty," Kenya Times" (June 29,
1989).
Hardin, Garrett, "'The tragedy of the commons," in
Garrett Hardin and John Baden (Eds.), Managing
the Commons (San Francisco: W. H. Freeman and
Company, 1977) [originally published in 1968].
Hardin, Garrett, "'An operational analysis of
'responsibility'," in Garrett Hardin and John Baden
(Eds.), Managing the Commons (San Francisco,
1977).
Hardin, Garrett, "Denial and disguise," in Garrett
Hardin and John Baden (Eds.), Managing the Commons (San Francisco, 1977).
Haugerud, Angelique, "Land tenure and agrarian
change in Kenya," Africa, Vol. 59, No. 1 (1989), pp.
61-90.
Haugerud, Angelique, "The consequences of land
tenure reform among smallholders in the Kenya
Highlands," Rural Africana, Nos. 15-16 (Winter
1983), pp. 65-89.
Haugerud, Angclique, "Development and household
economy in two eco-zones of Embu District,"
Working Paper No. 382 (Nairobi: Institute of Development Studies, University of Nairobi, 1981).
Hjort, Anders, "Herds, trade, and grain: Pastoralism
in a rcgional perspective," in John Galaty, Dan
Aronson, Philip Salzman, and Amy Chouinard
(Eds.), The Future of Pastoral Peoples: Proceedings
of a Conference Held in Nairobi, Kenya, 4-8 August,
1980 (Ottawa: International Development Research
Center, 1981).
Hukkinen, Janne, Emery Roe and Gene Rochlin, "'A
salt on the land: A narrative analysis of the controversy over irrigation-related salinity and toxicity in
California's San Joaquin Valley," Policy Sciences
(forthcoming).
Hunt, Diana, The Impending Crisis in Kenya: The Case
for Land Reform (Aldershot, England: Gower Publishing Company, 1984).
Hyden, Goren, No Shortcuts to Progress: African
Development Management in Perspective (Berkeley:
University of California Press, 1983).
Johnston, Bruce, and William Clark, Redesigning
Rural Development: A Strategic Perspective (Baltimore: The Johns Hopkins University Press. 1982).
Kawamura, Itsuo, "Summary and conclusions," in
Dominick Salvatore (Ed.), African Development
Prospects: A Policy Modeling Approach (New York:
Taylor & Francis on behalf of the United Nations,
1989).
Korten, David, "Community organization and rural
development: A learning process approach," Public
Administration Review, Vol. 40, No. 5 (September/
October 1980), pp. 480-51l.
Lamb, David, The Africans (New York: Vintage,
1984).
Leonard, David, "Disintegrating agricultural development," Food Research Institute Studies', Vol. XIX,
No. 2 (1984), pp. 177-186.
McCay, Bonnie, and James Acheson (Eds.), The
Question of the Commons." The Culture and Ecology
of Communal Resources" (Tucson: The University of
Arizona Press, 1987).
Ministry of Lands and Settlement, "'Programme of
Work and Performance Targets for the Year 1988,"
Memo MLS 2/001 Vol. lI/(16) (Nairobi: January
1988).
DEVELOPMENT NARRATIVES
Moris, Jon, - Managing Induced Rural Development
(Bloomington: International Development Institute,
Indiana University, 1981).
Moils, Jon, and Derrick Thom, African Irrigation
Overview: Main Report, Water Management Synthesis Report 37 (Logan: Utah State University,
1987).
Mutiso, Gideon Cyrus, "Historical perspective, existing ASAL programmes and institutional analysis,"
Technical Paper No. 3 in Republic of Kenya --Arid
and Semi-Arid Lands (ASAL) Development Programme: Summary of Technical Reports on the Strategy, Policy and ASAL Development Programme
1989-1993 (Nairobi: IFAD/UNDP, 1988).
Njeru, E. H. N., "Land adjudication and its implications for the social organization of the Mbere,"
Research Paper No. 73 (Madison: Land Tenure
Center, University of Wisconsin, 1978).
Odell, Malcolm, and Marcia Odell, "Communal area
livestock development in Botswana: Lessons for the
World Bank's Third Livestock Development Project," Paper presented to the World Bank (Amesbury, MA: Synergy International, 1986).
Odingo, Richard, "The dynamics of land tenure reform
and of agrarian systems in Africa: Land tenure study
in the Nakuru, Kericho and Machakos Areas of the
Kenya Highlands" (Rome: FAO, 1985).
Oluoch-Kosura, Willis in collaboration with S. E.
Migot-Adholla, A World Bank-sponsored study on
the effect of land registration in four localities from
two districts in Kenya (Nairobi: forthcoming).
Omolehinwa, Eddy and Emery Roe, "Boom and bust
budgeting: Repetitive budgetary processes in
Nigeria, Kenya and Ghana," Public Budgeting and
Finance, Vol. 9, No. 2 (Summer 1989), pp. 43-65.
Panel on Common Property Resource Management,
Proceedings of the Conference on Common Property
Resource Management (Washington, DC: National
Academy Press on behalf of the Board on Science
and Technology for International Development,
Office of International Affairs, National Research
Council, 1986).
Perrow, Charles, Normal Accidents: Living With High
Risk Technologies (New York: Basic Books, 1984).
Peterson, Stephen, "Microcomputers and institutional
development: Emerging lessons from Kenya," Paper
prepared for the Harvard Institute of International
Development Conference on Economic Reform
(Marrakech, Morocco: October 1988).
Pinckney, Thomas, John Cohen, and David Leonard,
"Microcomputers and financial management in
development ministries: Experience from Kenya,"
Agricultural Administration, Vol. 14, No. 3 (1983),
pp. 151-167.
Prince, Gerald, A Dictionary of Narratology (Lincoln:
University of Nebraska Press, 1987).
Ramakrishnan, Subramaniam, "Issues in budgeting
and fiscal management in Sub-Saharan Africa,"
Discussion paper prepared for the Workshop on
Economic Reform in Africa sponsored by the University of Michigan's Center for Research on Economic Development and USAID (Nairobi: July
1989).
Republic of Botswana, White Paper of 1975: National
299
Policy on Tribal Grazing Land (Gaborone, Botswana: Government Printer, 1975).
Roe, Emery, "Nonsense, fate and policy analysis: The
case of animal rights and experimentation," Agriculture and Human Values, (forthcoming.a).
Roe, Emery, "Applied narrative analysis: The tangency of literary criticism, social science and policy
analysis," New Literary History, (forthcoming.b).
Roe, Emery, "Narrative analysis for the policy analyst:
A Case study of the 1980-1982 Medfly controversy in
California," Journal of Policy Analysis and Management, Vol. 8, No. 2 (Spring 1989a), pp. 251-273.
Roe, Emery, "Six myths about livestock rangeland
development South of the Sahara," Rangelands, Vol.
11, No. 2 (October 1989b), pp. 217-221.
Roe, Emery, "Folktale development," American
Journal of Semiotics, Vol. 6, No. 2/3 (1989c), pp.
277-289.
Roe, Emery, "Perspectives from organization theory
on irrigation bureaucracies in East and West Africa,"
in Norman Uphoff with Prithi Ramamurthy and Roy
Steiner, Improving Performance of Irrigation
Bureaucracies: Suggestions for Systematic Analysis
and Agency Recommendations (Ithaca, New York:
Cornell University, 1988a).
Roe, Emery, "Management crisis in the new class,"
Telos, Number 76 (Summer 1988b), pp. 134-143.
Roe, Emery, "Deconstructing budgets," Diacritics,
Vol. 18, No. 2 (Summer 1988c), pp. 61--68.
Roe, Emery, "Counting on comparative budgeting in
Africa," International Journal of Public Administration, Vol. 11, No. 3 (May 1988d), pp. 341-356.
Roe, Emery, "Individualism and community in
Africa?: The case of Botswana," The Journal of
Modern African Studies, Vol. 26, No. 2 (June 1988e),
pp. 347-350.
Roe, Emery, "The expatriate advisor as senior policy
analyst," Policy Studies Review, Vol. 7, No. 3
(Spring 1988f), pp. 519-536.
Roe, Emery, "Last chance at the African kraal:
Reviving livestock projects in Africa," in J. T.
O'Rourke (Ed.), Proceedings of the 1987 International Rangeland Development Symposium (Morrilton, Arizona: Winrock International Institute for
Agricultural Development, 1987a).
Roe, Emery, "'Lantern on the stern: Policy analysis,
historical research, and Pax Britannica in Africa,"
African Studies Review, Vol. 30, No. 1 (March
1987b), pp. 45--62.
Roe, Emery, "The ceiling as base: National budgeting
in Kenya," Public Budgeting and Finance, Vol. 6,
No. 2 (Summer 1986), pp. 87-103.
Roe, Emery, "Project appraisal: A venture capitalist
approach," Development Policy Review, Vol. 3, No.
2 (November 1985), pp. 208-221.
Roe, Emery, "Range conditions around water sources
in Botswana and Kenya," Rangelands, Vol. 6, No. 6
(December 1984), pp. 247-249.
Roe, Emery, and Louise Fortmann, Season and
Strategy: The Changing Organization of the Rural
Water Sector in Botswana, Special Series in Resource
Management (Ithaca, New York: Rural Development Committee, Cornell University, 1982).
Roe, Emery, and Louise Fortmann, Allocation of
300
WORLD DEVELOPMENT
Water Points at the Lands (Gaborone, Botswana:
Ministry of Local Government and Lands and the
Center for International Studies, Cornell University,
1981).
Roth, Paul, "How narratives explain," Social Research,
Vol. 56, No. 2 (Summer 1989), pp. 449-478.
Sandford, Stephen, Management of Pastoral Develop°
ment in the Third World (Chichester, England: John
Wiley & Sons, 1983).
Schick, Allen, "Microbudgetary adaptations to fiscal
stress in industrialized democracies," Public Administration Review, Vol. 48, No. 1 (January/February
1988), pp. 523-533.
Shipton, Parker, Land, Credit and Crop Transitions
in Kenya: The Luo Response to Directed Development in Nyanza Province, PhD dissertation (Cambridge: University of Cambridge, 1985).
Sidahmed, Ahmed, and L. Koong, "'Application of
systems analysis to nomadic livestock production in
the Sudan," in James Simpson and Phylo Evangelou
(Eds.), Livestock Development in Subsaharan
Africa: CorL~traints, Prospects, Policy (Boulder,
Colorado: Westview Press, 1984).
Sorrenson, M. P. K., Land Reform in the Kikuyu
Country: A Study in Government Policy (Nairobi and
London: Oxford University Press, 1967).
Sullivan, Timothy, "Knowledge and method in the
study of public management," Paper prepared at the
Graduate School of Public Policy (Berkeley: University of California, 1987).
United Nations, "Africa recovery news feature - - the
african alternative: Policy recommendations" (New
York: United Nations, 1989).
United Nations Economic Commission for Africa,
African Alternative Framework to Structural Adjustment Programmes for Socio-Economic Recovery and
Transformation,
E/ECA/CM. 15/6/Rev.3
(Addis
Ababa: ECA 1989).
USAID, "Suggestions for the improvement of rangeland livestock projects in Africa: A panel report"
(Washington, DC: USAID, 1985).
Wangari, Esther, PhD dissertation on land registration
in Lower Embu District (New York: New School of
Social Research, forthcoming).
Wescott, Clay, "Microcomputers for improved
budgeting by the Kenya government," Development
Discussion Paper No. 227 (Cambridge, MA:
Harvard Institute for International Development,
1986).
Wildavsky, Aaron, The Politics of the Budgetary
Process, 4th edition (Boston: Little, Brown and
Company, 1984).
Wilson, Rodney, "'The Economic implications of land
registration in Kenya's smallholder areas," Staff
Working Paper No. 91 (Nairobi: Institute of
Development Studies, University of Nairobi, 1971).
World Bank, World Development Report 1988 (New
York: Oxford University Press, 1988.
World Bank, Kenya: Agricultural Credit Policy Review,
Report No. 5619-KE (Washington, DC: August
1985).